Deteksi Loneliness State Berbasis Chatbot Menggunakan Deep Learning

Rahayu, Ririn Tri (2023) Deteksi Loneliness State Berbasis Chatbot Menggunakan Deep Learning. Masters thesis, Institut Teknologi Sepuluh Nopember Surabaya.

Text 07111950052001-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2025. Download (2MB)

Abstract

Rasa kesepian yang dirasakan dan isolasi sosial telah mengalami peningkatan selama dekade terakhir, terutama pada negara-negara yang memiliki populasi masyarakat dewasa dan lansia yang tinggi. Keadaan tersebut diperparah karena adanya kebijakan yang harus diambil untuk menghadapi wabah COVID-19 selama dua tahun terakhir. Dengan menggunakan pendekatan metode Natural Language Processing (NLP) untuk mengukur sentimen dan variabel yang menandakan kesepian dalam transkrip teks kalimat yang diucapkan responden, penelitian ini memanfaatkan penggunakan teknologi deep learning dalam evaluasi wawancara tentang kesepian. Deteksi kesepian menggunakan metode Deep Neural Network (DNN) dan Long Short-Term Memory (LSTM). Responden yang kesepian dan yang tidak mengalami kesepian dibandingkan (menggunakan ukuran kualitatif dan kuantitatif). Individu yang lebih kesepian (yang telah ditentukan berdasarkan ukuran kualitatif) membutuhkan waktu lebih lama untuk menjawab pertanyaan tentang kesepian mereka dan mengungkapkan lebih banyak kesedihan dalam menjawab. Responden wanita lebih banyak mengakui rasa kesepian yang dialaminya pada saat proses wawancara kualitatif dibandingkan dengan responden pria. Saat merespon wawancara, pria lebih cenderung menghasilkan ekspresi ketakutan dan kebahagiaan. Saat dilakukan pelatihan data menggunakan data wicara model DNN yang dibangun memberikan hasil akurasi 99,61% dalam memprediksi kesepian
=====================================================================================================================================
Perceived loneliness and social isolation have been on the rise over the past decade, especially in countrises with rapidly aging populations and, most notably, as a result of the years. By using a natural language processing (NLP) approach to quantify sentiment and variables tha signal loneliness in transcribed spoken text of oleder persons, this paper investigates the use of deel learning technology in the evaluation of interviews on loneliness. We conducted loneliness state detection using Deep Neural Network (DNN) and Long Short-Term Memory (LSTM). Participans who were lonely and those who weren’t were compared (using both qualitative and quantitative measures). Individuals who were lonelier (as determined by qualitative measures) took longer to respond to questions about their loneliness and expressed more grief in their answers. When asked about loneliness, more women than men admitted it during the qualitative interview. When responding, men were more likely to utilize expressions of dread and happiness. When trained on speech data, DNN models were 99,6% accurate at predicting loneliness

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	identifikasi kesepian, klasifikasi biner, deep learning, DNN loneliness identification, binary classification, deep learning, DNN
Subjects:	T Technology > T Technology (General) > T57.5 Data Processing
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User:	Ririn Tri Rahayu
Date Deposited:	04 Feb 2023 10:17
Last Modified:	04 Feb 2023 10:17
URI:	http://repository.its.ac.id/id/eprint/96177

Actions (login required)

View Item